Re: Direct IO

From: Lucille Williams <luluwilliams_at_earthlink.net>
Date: Sun, 24 Jun 2001 12:15:28 GMT
Message-ID: <3b35d855.55967894@news.earthlink.net>

On Sat, 23 Jun 2001 00:45:38 GMT, "Dave Haas" <davidh_at_no.spam.hotmail.com> wrote:

Direct I/O like AIO, KAIO, and List I/O are provided as access methods on most UNIX machines. Its up to Oracle to work with the Veritas's and Sun's of the world to know how to implement these features. I have head that when Oracle writes to a datafile on UFS or VXFS it WRITE()'s with the osync flag on which bypasses cache.
>Hi.
>
>I have a question with regard to terminology. In several posts (and an
>argument I had with Howard a while ago :) people have used the term 'Direct
>IO'. To be perfectly honest I'm not exactly sure what that means. AFAIK
>the IO options are 1) file-system buffered and 2) Raw IO. I heard (or more
>to the point, read) a post that said something to the effect of '... direct
>I/O means that the buffer cache is not involved in the operation ...'. Does
>that have something to do with the sort-direct-write operation and the old
>SORT_WRITE_BUFFERS and SORT_WRITE_BUFFER_SIZE? I'm a little confused (as
>usual) ...
>
>Dave
>
>
Received on Sun Jun 24 2001 - 07:15:28 CDT